Handheld electronic devices and other portable electronic devices are becoming increasingly popular. Examples of handheld devices include handheld computers, cellular telephones, media players, and hybrid devices that include the functionality of multiple devices of this type. Popular portable electronic devices that are somewhat larger than traditional handheld electronic devices include laptop computers and tablet computers.
Handheld wireless communications devices often have several functions that involve digital audio signal processing. For example, consider their use as a mobile telephony device (e.g., a cellular telephone handset). Following a call set up or connection phase, a simultaneous two-way voice conversation between a local user of the device and another (remote) user in a telephone call may be enabled as follows.
A so-called uplink chain in the device is responsible for digitizing the local user's speech that has been detected by a built-in microphone. This may result in a raw digital audio signal or bit stream, e.g. a pulse code modulated, PCM, audio signal or bitstream. The uplink chain then digitally codes the raw signal, to remove its redundant content. For instance, a 64 kbits/sec raw speech bitstream may be encoded as a 14 kbits/sec bitstream, without causing a noticeable drop in sound quality. Next, the uplink chain modulates a RF carrier signal with the coded signal (and other information regarding the call). An antenna of the device is then driven with the modulated RF carrier. The local user's speech is thus transmitted to the cellular telephone network.
To enable the above-mentioned two-way conversation, a downlink chain is provided in the device, so that the local user can hear the remote user's speech. The downlink chain operates in parallel with or simultaneously as the uplink chain, to enable the real-time two-way conversation. The downlink chain may essentially perform the reverse of the uplink chain's operations. Thus, an antenna of the device outputs a downlink RF signal sent by the cellular telephone network. The downlink chain then demodulates the downlink RF signal to yield a so-called baseband signal. The latter contains a coded audio signal, which includes an encoded version of the captured speech of the remote user. The coded audio signal is decoded (e.g., into a PCM bitstream), converted to analog format and then played to the local user, through a receiver or speaker of the device. To render higher quality or better sound when an audio signal is played back, various signal processing operations may be performed on the digital audio signal in both the downlink and uplink chains. These may include noise filtering or noise suppression (sometimes referred to as noise cancellation), gain control, and echo cancellation.
Most handheld wireless communications devices are typically certified for use with a given cellular communications network. This may be in accordance with a specification that is governed by an approved authority such as the PCS Type Certification Review Board (PTCRB). The certification process entails the laboratory testing of a manufactured specimen of the device, to determine its compliance with the specification. For example, the audio portion of the specification for Global System for Mobile communications, GSM, devices requires that an artificial speech signal (or “cert signal”) be sent over the air during a wireless call with the device. For the downlink portion of the test, the cert signal is received over the air by the device. The sound pressure level and spectrum of this cert signal as output by the device's receiver (earpiece speaker) is then measured, at a given volume or loudness setting of the device. For the uplink portion of the test, the cert signal is transmitted by the device over the air to a receiving test station where it is converted into sound. In order for the device to pass the certification test, the measured sound output (which is a function of frequency) needs to fall within a certain spectral and sound pressure level range (also referred to as a “mask”) that is defined in the specification, for the duration of the signal (e.g., about twenty seconds).